ISSN 0327-9383
Mastozoología Neotropical; 5(2):109-116
SAREM, 1998
MEDULLAR TYPE AND CUTICULAR SCALE
PATTERNS OF HAIRS OF RODENTS AND SMALL
MARSUPIALS FROM THE MONTE SCRUBLAND
(SAN LUIS PROVINCE, ARGENTINA)
Gustavo J. Fernández* and Silvia M. Rossi
Area de Zoología, Facultad de Química, Bioquímica y Farmacia, Universidad Nacional de San Luis.
Chacabuco y Pedernera (5700). San Luis. Provincia de San Luis. Argentina.*Present address: Instituto de Biología y Medicina Experimental (CONICET). Vuelta de Obligado 2490. (1428) Capital
Federal. Argentina. E-mail mermoz@proteus.dna.uba.ar
ABSTRACT: Hair characteristics are often used to identify mammalian prey in guts, feces
or pellets of predators when other more easily identifiable remains, such as bone fragments
or teeth are not available. Indeed, hair can provide qualitative data about prey species. By
using guard hair cuticular scale patterns and medullar types, we built a key which enabled
us to accurately identify nine of thirteen rodents and small marsupials found in San Luis
Province (Argentina). Our results support the use of hairs to identify mammalian prey but
emphasize that at least two criteria must be examined to accurately distinguish among
species.
Key words: hairs, guard hairs, medullar type, cuticular scale pattern, hair identification,
rodents, Monte scrubland, San Luis.
INTRODUCTION
Several studies analyzing mammalian hair
structure have shown that it is generic and
possesses species-specific characteristics. Hair
analysis allows the identification of prey items
from remains in feces, guts or pellets when
other identifiable remains such as teeth or
bones are missing or too fragmented (Day,
1966; Gurini, 1985; Capurro et al., 1988).
Excessive fragmentation of remains is very
common where small carnivores (Day, 1966;
Gurini, 1985) and scavengers (Capurro et al.,
1988) are involved.
The fur of most mammalian species is composed by two kinds of hairs. Guard hairs which
are long, straight and heavily pigmented, and
fine hairs which are shorter and less pigmented
than the former. Both show three layers of
keratin: an external cuticle surrounding the hair,
a cortex, and an internal layer named medulla,
which is made up of loosely packed cells, often
Recibido 10 febrero 1997. Aceptado 20 junio 1998.
containing air. Only guard hairs show cuticular and medullar patterns consistent enough to
allow reliable species identification (Day, 1966;
Gurini, 1985).
Cuticular scale patterns, types of medulla and
pigmentation and shape are the most useful
features for identifying mammalian hairs (Day,
1966; Gurini, 1985; Busch, 1986; Capurro et
al., 1988; Chehébar and Martín, 1989). Indeed,
such characteristics do not vary with sex, diet
or age (Gurini, 1985). Hence they have an
important diagnostic value as systematic or
phylogenetic elements as well as for predator
diet analysis.
Several authors have attempted to develope
identification keys for mammalian prey in
Argentina based on some of their hair features. For instance, based on cuticular scale
patterns only, Piantanida and Petriella (1976)
unsuccessfully tried to build a species specific
hair identification key for cricetids of Buenos
Aires. Using cuticular scale patterns and med-
G.J. Fernández y S.M. Rossi
110
ullar types Busch (1986) and Capurro et al.
(1988) found that these two hair features were
satisfactory for the identification of some
mammalian prey. In turn, Chehébar and Martín
(1989) suggested that the same two hair characteristics were not enough to identify all
mammalian species from Patagonia.
In this report, we study the differences in
cuticular scale patterns and types of medulla
for twelve species of rodents and one small
marsupial (Table 1) from the north and central region of San Luis Province (Monte scrubland). Using these differences we built a hair
identification key for the most common predators’ food items within this region.
METHODS
We used only guard hairs, because they show constant features for any species. Fine hairs, instead,
are less constant and have simplified versions of
those features found in guard hairs (Day, 1966).
Further, the digestive process frequently destroys
fine hairs, whereas guard hairs are remarkably more
resistant.
Mammalian guard hairs usually have a narrow
proximal region, followed by a broad shield section, which narrows towards the distal portion of
the hair. The nomenclature for cuticular scale patterns and medulla types follows Hausman (1924,
1932), and later modifications by Benedict (1957)
and Gurini (1985).
We took hairs from the head, neck, back and
belly of the species listed in Table 1. Sampled
animals came from the Zoology Area Mammalian
Collection of San Luis University (Argentina), and
from live animals kept at the Laboratory of Ecophysiology (San Luis University, Argentina).
Hairs sampled were rinsed in a sulfuric acid-alcohol 70% solution and were mounted on microscope slides. The medulla was observed throughout. When the medulla could not be identified easily, we cleared the hair with peroxide (Gurini, 1985).
In order to identify the cuticular scale pattern, once
identified the medullar type, we made a cast of
each hair on a thin layer of correction fluid which
had been previously extended on a microscope slide
(Schemnitz, 1980). Hairs were laid on this film,
allowing for the correction fluid to dry. Then, hairs
were peeled off, leaving behind a cast of the scales.
Hairs were observed through a microscope at 100x
and 400x.
RESULTS
Guard hairs of different body areas showed
coloration, length and diameter differences but
similar cuticular scale patterns and medullar
types for the same individual. Further, hair
structure was similar for individuals of the same
species, but differences were detected between
species, enabling us to identify most of the
species studied. Cuticular scale pattern and
medullar type of each species are described
below.
Cuticular scale patterns
Hair cuticular scale patterning from the different species were similar at the shield area, but
major variation was noted at the proximate
and distal portions of them. The different cuticular scale patterns observed are shown in
Fig. 1.
The cuticular scales of tip and shield portions of M. pusilla’s hairs showed a mosaiclike pattern. Their basal portion, however, presented a pectinate pattern which merged into
diamond petal in the proximal parts of the shaft.
Ctenomys’s hairs had only the basal and
mean portions pigmented. Guard hairs showed
a mosaic scale pattern at the shield, turning
into a wavy crenated one at the top of the
shield and at the tip. At the basal portion, the
scales appeared to be pointing backwards
showing an oblique mosaic pattern.
Like Ctenomys sp., Microcavia australis
showed an oblique mosaic pattern on the basal
portion, turning to mosaic at the shield, and to
a wavy crenated pattern at the tip. Their hairs,
although shorter than Ctenomys’s, are heavily
pigmented. Lagostomus maximus and
Pediolagus salinicola showed similar cuticular scale patterns for the shield and tip regions,
but their basal portion had a regular polygonal
pattern. Pigment in Lagostomus’s hairs was
concentrated in the medial zone, whereas
Pediolagus showed a more uniform pigmentation.
Cricetid species also presented uniformly
pigmented hairs. Hairs of Eligmodontia typus
and Graomys griseoflavus appeared to be identical. Guard hair of Graomys were slightly
longer than those of Eligmodontia, but their
HAIR CHARACTERISTICS OF SMALL MAMMALS OF SAN LUIS
111
Table 1. List of species, number of specimens and total samples taken used for hair analysis.
Species
Lagostomus maximus
Pediolagus salinicola
Microcavia australis
Ctenomys sp.
Phyllotis darwini
Graomys griseoflavus
Eligmodontia typus
Akodon molinae
Calomys laucha
Calomys venustus
Calomys musculinus
Marmosa pusilla
Order, family
Rodentia, Chinchillidae
Rodentia, Caviidae
Rodentia, Caviidae
Rodentia, Ctenomyidae
Rodentia, Cricetidae
Rodentia, Cricetidae
Rodentia, Cricetidae
Rodentia, Cricetidae
Rodentia, Cricetidae
Rodentia, Cricetidae
Rodentia, Cricetidae
Marsupialia, Didelphyidae
cuticular scale patterns were similar. The proximal portion of hairs had a lanceolate pattern
with the shield and tip portions displaying a
mosaic-like pattern. E. typus occasionally
showed, in the basal portion of the proximal
region, a coronal sheathed cuticular scale pattern. This pattern was not observed in
Graomys’s hairs, but it was restricted to the
narrowest part of Eligmodontia’s hairs. However, less than 10% of its hair presented this
pattern, and so it is possible that sometimes
identification may be hindered as this region
could be missing.
Calomys musculinus and C. laucha presented
mosaic scale patterns at the shield and tip region of hairs. In the proximate portion of hairs
C. musculinus had a pectinate scale pattern,
whereas C. laucha had a lanceolate one, which
turned into a chevron-like pattern towards the
shield area. Calomys venustus and C. callosus
presented a similar cuticular scale pattern at
the shield and tip portions of hairs. At the
shield, they have a mosaic-like pattern which
turned into a wavy crenated one at the tip.
However, in C. venustus the mosaic pattern
was more reduced than in C. callosus. Proximate regions of hairs of these species were
lanceolate in the latter and pectinate in the
former.
Akodon molinae showed a hair cuticular scale
pattern similar to that of C. callosus. However, as observed in C. venustus, shield mosaic patterns were shorter than those present
Number of individuals
Sample size
2
1
3
2
2
6
8
8
6
6
10
2
12
8
18
14
14
14
24
24
20
20
26
12
in C. callosus. Phyllotis darwini’s hairs showed
a lanceolate pattern at the proximate region
which turned into mosaic at the shield area
and to wavy crenated at the tip.
Medullar pattern
In almost all the species, medullar types varied from cells in an uniseriate (i.e. in a single
row) to multiseriate arrangement (i.e. in two
or more parallel rows), depending on hair diameter (Fig. 2). For most species, proximal
and distal regions of hairs frequently had an
uniseriate pattern. Only Ctenomys’s hairs had
a fragmented medulla at the tip. On the other
hand, shield regions of hairs showed the highest interspecific pattern variation. It was insufficient, however, to separate most species.
Almost all cricetids presented a triseriate
medulla at the shield region (Eligmodontia
typus, Calomys musculinus, C. laucha, C.
venustus, C. callosus, Akodon molinae and
Graomys griseoflavus). Only Phyllotis darwini
had a multiseriate medullar type. On the other
hand, Ctenomys sp. and Lagostomus maximus
had hairs with a lattice-like medulla at the
shield areas, whereas Pediolagus salinicola
showed a reticulate medullar pattern.
Microcavia australis showed heavily pigmented hairs, instead of the peroxide-washing
process. It made identification difficult, but it
appears to be of lattice type.
In Marmosa pusilla, shield portions were
poorly defined and there was little difference
112
Fig. 1.
G.J. Fernández y S.M. Rossi
Cuticular scale patterns of some small mammals from the Monte scrubland (San Luis Province, Argentina). a)
Mosaic (Microcavia australis); b) wavy crenated (Microcavia australis); c) chevron (Calomys laucha); d)
coronal (Calomys laucha); e) pectinate (Calomys musculinus); f) and g) lanceolate (Eligmodontia typus and
Graomys griseoflavus, respectively); h) diamond petal (Marmosa pusilla); i) regular polygonal (Pediolagus
salinicola). a), b) scale 100x (1 cm: 2.5 mm); c)-i) scale 400x (1 cm: 0.08 mm).
HAIR CHARACTERISTICS OF SMALL MAMMALS OF SAN LUIS
Fig. 2.
113
Medullar types of some small mammals from the Monte scrubland (San Luis Province, Argentina). a) uniseriate
(Marmosa pusilla, shield region); b) uniseriate (Calomys laucha, basal portion); c) triseriate (Eligmodontia
typus); d) multiseriate (Ctenomys sp.); e) fragmented (Ctenomys sp.); f) lattice (Lagostomus maximus); g)
triseriate (Graomys griseoflavus); h) multiseriate (Phyllotis darwini). a)-c), g) and h) scale 400x (1 cm: 0.08
mm); d)-f) scale 100x (1 cm: 2.5 mm).
G.J. Fernández y S.M. Rossi
114
between guard and fine hairs. Hair medulla in
this species was totally uniseriate. Interestingly,
Gurini (1985) found similar medullar types in
other marsupials (Didelphis albiventris,
Lutreolina crassicaudata, and Marmosa agilis).
Hair identification key
The differences in cuticular scale pattern and
medullar type found among the studied species, allowed us to build a key for species
identification. Since medullar patterns showed
to be similar at the extremes of the hair, we
first used the patterns found at the shield area.
We also used the differences in cuticular scale
pattern at the proximate and distal regions of
hairs. We discarded those cuticular patterns
observed at the shield regions because they were
similar for almost all the analyzed species.
The analysis of these hair features allowed
us to clearly identify 9 of the 13 species studied. For the rest of them, hair features differed
too little for unambiguous identification.
Hair Identification Key for Rodents and Small Marsupials
from the Monte Scrubland (San Luis, Argentina)
1
1'
Hairs without visible medulla .............................................................................. Microcavia australis
Hairs with visible medulla ................................................................................................................... 2
2
2'
Hairs with uni, bi or triseriate medulla .............................................................................................. 3
Hairs with other medullar type ........................................................................................................... 8
3
3'
Hairs with uniseriate medulla .................................................................................... Marmosa pusilla
Hairs with bi or triseriate medulla ...................................................................................................... 4
4
4'
Proximal portion of the hair with lanceolate cuticular scales ........................................................... 5
Proximal portion of the hair with another cuticular scales pattern .................................................. 6
5
Distal portion of the hair with wavy crenated cuticular scale pattern .................... Akodon molinae
Calomys callosus
Distal portion of the hair with mosaic cuticular scale patter ................................... Calomys laucha
Graomys griseoflavus
5'
6
6'
Proximal portion with coronal cuticular scale pattern ......................................... Eligmodontia typus
Proximal portion of the hair with pectinate cuticular scale pattern .................................................. 7
7
7'
Distal portion of the hair with wavy crenated cuticular scale pattern .................. Calomys venustus
Distal portion of the hair with mosaic cuticular scale pattern .......................... Calomys musculinus
8
8'
Hairs with multiseriate medulla ................................................................................. Phyllotis darwini
Hairs with other medulla type ............................................................................................................. 9
9
9'
Hairs with lattice medulla .................................................................................................................. 10
Hairs with reticulate medulla ............................................................................. Pediolagus salinicola
10
10'
Distal portion of the hair with discontinuous medulla (fragmented type) ................... Ctenomys sp.
Distal portion of the hair with medulla with uniseriate pattern ...................... Lagostomus maximus
HAIR CHARACTERISTICS OF SMALL MAMMALS OF SAN LUIS
DISCUSSION
The cuticular scale patterns and types of medulla found in hairs of each of the studied
species are similar to those reported in previous works in different regions (Busch, 1986;
Capurro et al., 1988; Chehébar and Martín,
1989).
Types of medulla are related to hair diameter as suggested by Hausman (1932). Finer
hairs can show an uniseriate pattern of cells in
the medulla or even the total absence of them.
The former is the case for Marmosa pusilla.
This species shows the finest guard hair, with
a very simple type of medulla (uniseriate).
Cricetid rodents, instead, show medullar patterns which are frequently triseriate at the shield
region and uniseriate at the narrowest extremes
of hairs. Larger rodents have thicker guard
hairs, with more complex medullar patterns.
However, M. pusilla’s hairs show a medullar
pattern similar to that of Lutreolina
crassicaudata’s and Didelphis albiventris’s
(Didelphidae) hairs, two larger marsupials
(even similar to some of the larger rodents
analyzed here), which is formed by only one
cell series (Gurini, 1985). In these marsupials,
hair diameters are very different ( 27 - 31μ for
M. pusilla; 51 - 61μ for L. crassicaudata; and
75 - 109μ for D. albiventris), but this has no
effect on the medullar pattern. It is possible to
consider that the uniseriate medullar pattern is
a characteristic of the subclass.
Cuticular scale patterns can also be related
to hair diameter. At the shield region, in almost all species, it shows a mosaic-like or wavy
crenated scale pattern, whereas at the proximate region, it can become lanceolate, pectinate or show a regular polygonal scale distribution. However, the relationship between diameter and scale pattern is no so clear at the
tip. Scale patterns are, in this case, related to
the pattern present at the shield rather than to
hair diameter.
In a previous work, Chehébar and Martín
(1989) used mainly the proximal portion of
hairs for identification of mammals at genus
level, and only in few cases it was usefully for
species identification. Accordingly, we must
resort to the rest of the hair (i.e., the distal
115
portion) for the proper identification of the
species here studied.
As Gurini (1985) has noted, species identification based on hair features is difficult within
the Cricetidae family because there are only
slight interspecific differences. We were not
able to distinguish from one another, only two
pairs of cricetids when using both the cuticular scale patterns and medullar types (Calomys
laucha and Graomys griseoflavus, and Calomys
callosus and Akodon molinae). In these cases,
the species have hairs with a lanceolate scale
pattern at proximate region, but A. molinae
and C. laucha show sometimes a chevron
pattern at the broadest region of the shield area
which could enable their identification. These
results are in accordance with those of Bush
(1986) for cricetines from the Pampa region.
The use of this technique shows to be efficient enough to identify most mammalian prey
species studied from the Monte region. It is
possible that other hair features (e.g.: crosssections) could yield more accurate identifications. However, this identification is eminently
qualitative in content, fact which renders any
attempt for quantification, not enabling us to
tackle an estimate of prey numbers in a predators diet. On the other hand, phylogenetic or
systematic implications of hair characteristics
remain still unclear and need further study.
ACKNOWLEDGEMENTS
We thank Ana M. Brigada for her support during this
study as well as facilitating the access to the Zoology
Area Mammalian Collection of the San Luis University,
Argentina). We also thank Enrique Caviedes-Vidal and
the Ecophysiology Laboratory (San Luis University,
Argentina) for providing hairs of live animals. Juan C.
Corley, Angel F. Capurro and Myriam E. Mermoz critically read the manuscript, and made constructive comments. We thank also C. Chehébar and an anonymous
reviewer for their valuable suggestions. Finally, we
are grateful to the Zoology Area staff for their logistic support.
LITERATURE CITED
BENEDICT, F.A. 1957. Hair structure as a generic
character in bats. University of California
Publications. Zoology, 59:285-584.
BUSCH, M. 1986. Identificación de algunas especies de
pequeños mamíferos de la Provincia de Buenos Aires mediante características de sus pelos. Physis
44:113-118.
CAPURRO, A.; A. TRAVAINI, and A. NOVARO. 1988.
Identificación de pequeños mamíferos de la Provincia de Neuquén a través de características de sus
pelos. (V Reunión de la Sociedad Argentina de
Mastozoología, Tucumán, Noviembre de 1988).
CHEHÉBAR, C. and S.MARTIN. 1989. Guía para el
reconocimiento microscópico de los pelos de los
mamíferos de la Patagonia. Acta Vertebrata (Doñana),
16:247-291.
DAY, M.G. 1966. Identification of hair and feather
remains in the gut and faeces of stoat and weasels.
Journal of Zoology (London), 148:201-217.
GURINI, L.B. 1985. Valor diagnóstico del pelaje y su
aplicación al estudio al estudio de las interacciones
tróficas, con referencia a especies del Delta Bonaerense. 179 pp. Tesis Doctoral, Universidad Nacional
de La Plata.
HAUSMAN, L.A. 1924. Further studies of the rlationship
of the structural characters of mammalian hair.
American Naturalist, 58:544-557.
HAUSMAN, L.A. 1932. The cortical fusi in mammalian
hair shafts. American Naturalist, 66:461-470.
PIANTANIDA, M. and A.M. PETRIELLA. 1976. Estudio morfológico de algunas especies de roedores de
la Provincia de Buenos Aires con el microscopio
electrónico de barrido. Physis, 35:105-124.
SCHEMNITZ, S. 1980. Wildlife management techniques.
The Wildlife Society, Washington D.C. 4th edition.